Guardado en:
Detalles Bibliográficos
Autores principales: Nakayama, Shogo, Okuda, Masahiro
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:https://arxiv.org/abs/2601.04555
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Tabla de Contenidos:
  • Conventional semi-supervised contrastive learning methods assign pseudo-labels only to samples whose highest predicted class probability exceeds a predefined threshold, and then perform supervised contrastive learning using those selected samples. In this study, we propose a novel loss function that estimates the confidence of each sample based on the entropy of its predicted probability distribution and applies confidence-based adaptive weighting. This approach enables pseudo-label assignment even to samples that were previously excluded from training and facilitates contrastive learning that accounts for the confidence of both anchor and positive samples in a more principled manner. Experimental results demonstrate that the proposed method improves classification accuracy and achieves more stable learning performance even under low-label conditions.